LCA Methodology
Cumulative Exergy Demand
LCA Methodology
Applying Cumulative Exergy Demand (CExD) Indicators to the ecoinvent Database* Michael E. Bösch1**, Stefanie Hellweg1, Mark A.J. Huijbregts2 and Rolf Frischknecht3 1 Institute
of Environmental Engineering, ETH Zürich, 8093 Zürich, Switzerland of Environmental Science, Institute for Wetland and Water Research, Faculty of Science, Radboud University Nijmegen, P.O. Box 9010, NL-6500 GL Nijmegen, The Netherlands 3 ecoinvent Centre, Empa, Ueberlandstrasse 129, 8600 Duebendorf, Switzerland 2 Department
** Corresponding author (
[email protected]) DOI: http://dx.doi.org/10.1065/lca2006.11.282 Please cite this paper as: Bösch ME, Hellweg S, Huijbregts MAJ, Frischknecht R (2007): Applying Cumulative Energy Demand (CExD) Indicators to the ecoinvent Database. Int J LCA 12 (3) 181–190 Abstract Goal, Scope and Background. Exergy has been put forward as an indicator for the energetic quality of resources. The exergy of a resource accounts for the minimal work necessary to form the resource or for the maximally obtainable amount of work when bringing the resource's components to their most common state in the natural environment. Exergy measures are traditionally applied to assess energy efficiency, regarding the exergy losses in a process system. However, the measure can be utilised as an indicator of resource quality demand when considering the specific resources that contain the exergy. Such an exergy measure indicates the required resources and assesses the total exergy removal from nature in order to provide a product, process or service. In the current work, the exergy concept is combined with a large number of life cycle inventory datasets available with ecoinvent data v1.2. The goal was, first, to provide an additional impact category indicator to Life-Cycle Assessment practitioners. Second, this work aims at making a large source of exergy scores available to scientific communities that apply exergy as a primary indicator for energy efficiency and resource quality demand. Methods. The indicator Cumulative Exergy Demand (CExD) is introduced to depict total exergy removal from nature to provide a product, summing up the exergy of all resources required. CExD assesses the quality of energy demand and includes the exergy of energy carriers as well as of non-energetic materials. In the current paper, the exergy concept was applied to the resources contained in the ecoinvent database, considering chemical, kinetic, hydro-potential, nuclear, solar-radiative and thermal exergies. The impact category indicator is grouped into the eight resource categories fossil, nuclear, hydropower, biomass, other renewables, water, minerals, and metals. Exergy characterization factors for 112 different resources were included in the calculations. Results. CExD was calculated for 2630 ecoinvent product and process systems. The results are presented as average values and for 26 specific groups containing 1197 products, processes and infrastructure units. Depending on the process/product group considered, energetic resources make up between 9% and 100% of the total CExD, with an average contribution of 88%. The exergy of water contributes on the average to 8% the total exergy demand, but to more than 90% in specific process groups. The average contribution of minerals and metal ores is 4%, but shows an average value as high as 38% and 13%, in metallic products and in building materials,
respectively. Looking at individual processes, the contribution of the resource categories varies substantially from these average product group values. In comparison to Cumulative Energy Demand (CED) and the abiotic-resource-depletion category of CML 2001 (CML'01), non-energetic resources tend to be weighted more strongly by the CExD method. Discussion. Energy and matter used in a society are not destroyed but only transformed. What is consumed and eventually depleted is usable energy and usable matter. Exergy is a measure of such useful energy. Therefore, CExD is a suitable energy based indicator for the quality of resources that are removed from nature. Similar to CED, CExD assesses energy use, but regards the quality of the energy and incorporates non-energetic materials like minerals and metals. However, it can be observed for non-renewable energy-intensive products that CExD is very similar to CED. Since CExD considers energetic and non-energetic resources on the basis of exhaustible exergy, the measure is comparable to resource indicators like the resource use category of Eco-indicator 99 and the resource depletion category of CML 2001. An advantage of CExD in comparison to these methods is that exergy is an inherent property of the resource. Therefore less assumptions and subjective choices need to be made in setting up characterization factors. However, CExD does not cover societal demand (distinguishing between basic demand and luxury), availability or scarcity of the resource. As a consequence of the different weighting approach, CExD may differ considerably from the resource category indicators in Eco-indicator 99 and CML 2001. Conclusions. The current work shows that the exergy concept can be operationalised in product life cycle assessments. CExD is a suitable indicator to assess energy and resource demand. Due to the consideration of the quality of energy and the integration of nonenergetic resources, CExD is a more comprehensive indicator than the widely used CED. All of the eight CExD categories proposed are significant contributors to Cumulative Exergy Demand in at least one of the product groups analysed. In product or service assessments and comparative assertions, a careful and concious selection of the appropriate CExD-categories is required based on the energy and resource quality demand concept to be expressed by CExD. Recommendations and Perspectives. A differentiation between the exergy of fossil, nuclear, hydro-potential, biomass, other renewables, water and mineral/metal resources is recommended in order to obtain a more detailed picture of resource quality demand and to recognise trade-offs between resource use, for instance energetic and non-energetic raw materials, or nonrenewable and renewable energies. Keywords: Cumulative energy demand (CED); cumulative exergy demand (CExD); ecoinvent database; energy efficiency, exergy; life cycle assessment (LCA); resource efficiency, resource quality demand * ESS-Submission Editor: Dr. Gerald Rebitzer (Gerald.Rebitzer@ alcan.com)
Int J LCA 12 (3) 181 – 190 (2007) © 2007 ecomed publishers (Verlagsgruppe Hüthig Jehle Rehm GmbH), D-86899 Landsberg and Tokyo • Mumbai • Seoul • Melbourne • Paris
181
Cumulative Exergy Demand
LCA Methodology
Introduction
Exergy can be regarded as a measure of useful energy. While energy is motion or ability to produce motion, exergy is work (=ordered motion) or ability to produce work (Wall 1993). Hence exergy designates the quality or availability of energy. Energy can be transformed but is always conserved as states the first law of thermodynamics. Exergy in contrast, is consumed in all real world processes as entropy is produced, according to the second law of thermodynamics (Szargut 2005) (Eq. 1): δEx = T0∑∆S
(1)
Ex = Exergy (MJ) T0 = Temperature of the surroundings (K) S = Entropy (MJ/K) Since exergy is a property of both the system and the environment, a reference environment is required. While the ratio between exergy and energy of an energy source can be directly calculated within such a reference environment, further reference states need to be defined to calculate the exergy content of matter. Since the natural environment is not in thermodynamic equilibrium, reference species are required for all elements, representing the most stable compounds that are commonly occurring in the environment. The chemical exergy of a substance can be calculated by means of the formation reaction that contains only reference species and the substance itself. The exergy value characterizes the minimal amount of work that is required to form the substance, or reversely, the maximal amount of work that can be extracted from the substance in a specific environment. Hence exergy refers to the physical value of a substance (Wall 1993). All production and consumption processes require energetic and material resources from the environment that feature exergy. The exergy requirement of technical processes can thus be seen as an indicator of resource quality demand that weights each resource by its theoretical energetic usefulness. Szargut et al. (1988) presented a comprehensive methodology to calculate the exergy of energetic and non-energetic raw materials and provides an extensive list of exergy values for elements and various industrially used resources. This list of exergy values was updated in Szargut (2005). The present work refers to the reference species and resource exergies of Szargut (2005). However, some missing resources were adapted from Szargut et al. (1988). Data consistency is assured since only resources that contain elements with unaltered exergy values were adopted. Other authors have applied and further developed exergy demand as an indicator for energetic efficiency and resource quality demand. For instance, Finnveden et al. (1997) developed exergy-based characterization factors based on Swedish mineral resources to be used in the impact assessment of Life-Cycle Assessment (LCA). Exergy balances have also been developed for entire countries (Frank 1956) and for whole industries in order to estimate the exergetic efficiencies of rawmaterial conversion into products of economic value (Wall 1990, Wall et al. 1994, Schaeffer et al. 1992, Ozdogan et al. 1995). More recent studies focus on waste treatment (Dewulf
182
et al. 2000a, 2002a) and extend exergy analyses over the whole life cycle of products (Dewulf et al. 2000b, 2002b, 2004). In many of these studies it is stated that a database of exergy values for basic processes would be useful to fill data gaps and maintain data consistency. This emphasises the potential usefulness of coupling the extensive inventory work done in LCA with exergy indicators. The goals of the current paper are (i) to provide exergy scores for a large number of materials and processes and (ii) to compare the exergy scores with resource use and resource depletion scores from conventional LCIA methods. 1 1.1
Methods Cumulative exergy demand
In order to quantify the life cycle exergy demand of a product, the indicator Cumulative Exery Demand (CExD) is defined as the sum of exergy of all resources required to provide a process or product (Eq. 2). The notation CExD was chosen in this paper to stress the similarities to CED. CExD is equivalent to the definition of cumulative exergy consumption (CExC) of Szargut (2005), both quantifying the total exergy requirement of a product. Szargut et al. (1988) calculated CExC by adding up the total exergy requirement of a process over a time period (e.g. one year). The exergy requirement of one unit of process output was then obtained by dividing the total exergy requirement by the number of unit outputs during this time period. The emergence of large life cycle databases such as ecoinvent enables and facilitates a product-specific approach, since such databases provide the resource demand for each unit process. Hence, improved CExD scores can be calculated that indicate the exergy demand of a single product directly. CExD is specified in MJequivalents to highlight that it is an impact assessment indicator and not an inventory elementary flow. (2) CExD mi Ex(ch),i nj rex – e(k,p,n,r,t),i ch k p n r t
= cumulative exergy demand per unit of product or process (MJ-eq) = mass of material resource i (kg) = exergy per kg of substance i (MJ-eq/kg) = amount of energy from energy carrier j (MJ) = exergy to energy ratio of energy carrier j (MJeq/MJ) = chemical = kinetic = potential = nuclear = radiative = thermal exergy
Exergy is stored in resources in the form of chemical, thermal, kinetic, potential, nuclear and radiative energy. The assignment of the adequate type of exergy depends on resource use (Szargut 2005): • Chemical exergy is applied on all material resources, for biomass, water and fossil fuels (i.e. all materials that are not reference species in the reference state) • Thermal exergy is applied for geothermy, where heat is withdrawn without matter extraction Int J LCA 12 (3) 2007
LCA Methodology • • • •
Cumulative Exergy Demand
Kinetic exergy is applied on the kinetic energy in wind used to drive a wind generator Potential exergy is applied on potential energy in water used to run a hydroelectric plant Nuclear exergy is applied on nuclear fuel consumed in fission reactions Radiative exergy is applied on solar radiation impinging on solar panels
The above list refers only to resources that are included in the ecoinvent database. For instance, noise could also include kinetic energy, but it is not considered as a useful resource here. 1.2
Chemical exergy
The chemical exergy of a compound depends on its composition. If the composition is known, the chemical exergy can be calculated on the basis of the chemical formation reaction, utilising the Gibb's free energy of formation and the exergy of the chemical elements in the substance (Eq. 3). The exergy of chemical elements is provided by Szargut (2005) and data on ∆fGi° is available in handbooks (e.g. Atkins 2001). A solid material consisting of several chemical compounds is regarded as a mixture of separate grains (Finnveden et al. 1997). The chemical exergy is calculated by adding up the exergy of the mole fractions (Eq. 3). The standard chemical exergies for most of the ecoinvent resources have already been calculated in previous studies and are taken from Szargut et al. (1988, 2005). Molar chemical exergies exch can be approximated by the standard molar exergies ex°ch from the reference environment (Szargut 2005).
(3)
Exch nj ex°ch,j ∆fGi° ex°ch, el nel
: molar chemical exergy of material (kJ/mol) : mole fraction of substance j in material (–) : standard molar chemical exergy of substance (kJ/mol) : standard Gibb's free energy of formation of substance (kJ/mol) : standard partial molar chemical exergy of elements in substance (kJ/mol) : number of elements in compound (–)
The chemical composition of some resources, such as mineral substances (denoted as Minerals in this paper) is provided in ecoinvent data v1.2 (ecoinvent Centre 2005). Szargut (2005) provides exergy values for the majority of the minerals captured in the ecoinvent database. By contrast, other resources do not have a well-defined composition. For instance, rocks and ores are composed of various minerals. The composition may vary considerably in different locations even on a small geographical scale. In the case of rocks, exergy calculations were carried out for average rock compositions, assumed on the basis of information from external sources (see Table S1 for references: Supporting Information, online only). Ores are rocks that are mined for commercial metal extraction. Although the ecoinvent data
Int J LCA 12 (3) 2007
v1.2 includes metallic resources per kilogramme of pure metal only, the whole ore should be considered in order to account for the overall material extraction from nature (Finnveden et al. 1997). The ecoinvent elementary flow names provide the weight fraction of the specific metals for most of the ores. If the weight fraction was not specified, assumptions were made based on other information (see Table S1: Supporting Information, online only). The mineral composition of the ore is not provided by the ecoinvent names, because specific data is very scarce. Finnveden et al. (1997) performed exergy calculations of ores with data from mines that are mainly located in Scandinavia. Since the ecoinvent data v1.2 does not refer to the same mines, we did not directly use the data of Finnveden et al. (1997) for specific ores. There is no direct connection between the metal species and the composition of the ore (Lichtensteiger 2006). Hence, an average ore composition was assumed for all ores, represented by the median exergy value of the ores in Finnveden et al. (1997), which accounts for 0.63 MJ of exergy per kilogramme of ore. Differences in exergy values between the metals from ores are due to the metal concentration encountered in the ores. The higher the concentration of a metal in an ore, the smaller is the amount of ore to be mined in order to gain 1 kg of the metal. Therefore, metals that occur at high concentrations in ores require the extraction of less exergy than metals occurring at low concentrations. Hence, the decreasing concentration of metals in ores as a consequence of human extraction is reflected in the CExD factor for a resource. Several ores in ecoinvent data v1.2 contain two or more extractable metals. An allocation factor was applied to allocate the total exergy of an ore to the distinct metals. Allocation is necessary to avoid double counting. For instance, cadmium and zinc are retrieved from the same ore. Therefore, the exergy of the ore needs to be allocated to these two metals (Eq. 4). Allocation was performed by revenue whenever ecoinvent data v1.2 provides the revenue factors; otherwise, allocation by mass was applied (Althaus et al 2004, Althaus & Classen 2005) (see Table S1, Elements in ores: Supporting Information, online only). (4) Ex°ch,j Ex°ch,o cj a(r,m),j r m
= exergy per kg elementary metal j (MJ/kg) = exergy per kg ore (MJ/kg) = mass fraction of metal j in ore (–) = allocation factor for metal j (–) = revenue allocation = mass allocation
The chemical exergy value for water from Szargut (2005) was attributed to freshwater. Seawater is a reference species and does not feature exergy. Further reference species are the resources CO2, krypton, xenon in air, 'magnesium in water' (Szargut, 2005). 'Lithium in brine' is considered as Lithium in seawater and therefore as a reference species (see Table S1: Supporting Information, online only). For the five minerals borax, colemanite, stibnite, ulexite, and zirconia, no accurate data was found. No exergy scores were calculated for these minerals.
183
Cumulative Exergy Demand
LCA Methodology
Table 1: Assignation of exergy to the resource types of minerals, rocks, ores, and resources in air and water in the ecoinvent database v1.2 (key examples). The complete resource list and the corresponding references for resource composition are provided in Table S1 (Supporting Information, online only)
Unit
Composition
Ex°ch (MJ/unit)
Coefficient / Allocation
Ex°ch (MJ/unit product)
Comment
Minerals Anhydrite Sodium sulphate
kg kg
CaSO4 Na2SO4
0.06 0.15
1 1
0.06 0.15
– –
Rocks Granite
kg
51% albite, 25% quartz, 6% anortite, 10% biotite repl. by talc, 8% amphibole repr. by tremolite
0.068
1
0.068
Composition estimated
kg
Unknown
0.63
9.13
5.75
Exergy of average ore assumed; revenue allocation
kg
Unknown
0.63
54.05
34.05
Exergy of average ore assumed; revenue allocation
CO2 H2O H2O
0 50 0
– 1 –
0 50 0
Reference species in reference state – Reference species in reference state
Name
Elements in ores Ni, Ni 2.3E+0%, Pt 2.5E-4%, Pd 7.3E-4%, Rh 2.0E-5%, Cu 3.2E+0% in ore Ni, Ni 3.7E-2%, Pt 4.8E-4%, Pd 2.0E-4%, Rh 2.4E-5%, Cu 5.2E-2% in ore
Resources in air and water Carbon dioxide kg Freshwater m3 Seawater m3
Table 1 presents key examples of the exergy scores for the resource types of minerals, rocks, ores and resources in air and water. The complete resource list and the references are provided in Table S1 (Supporting Information, online only).
heat source is in general rather low, well below the temperature of the reference environment. Therefore no exergy is attributed to geothermal energy used in heat pumps.
Energetic resources in ecoinvent database v1.2 are either inventoried by mass, volume or energy content. For those inventoried by mass or volume, the energetic content per mass or volume is additionally provided. The exergy content is calculated through applying an exergy to gross calorific value ratio as provided by Szargut (2005). Assumptions are made for energy carriers where no ratio is provided (Table 2). The exergy to gross calorific value ratio for crude oil is estimated by the mean of the ratios for gasoline and lignite. The ratio for peat is estimated by the mean of the ratios for wood and lignite.
Kinetic exergy is equal to the kinetic energy, when the velocity is considered relative to the surface of the earth (Szargut 2005). Similarly, potential exergy is equal to the potential energy when it is evaluated with respect to the average level of the surface of the earth in the locality of the process under consideration (Szargut 2005). The energy of nuclear fuels is of very high quality because it intrinsically corresponds to a very high temperature. Hence, it can be assumed that the exergy of nuclear raw materials is equal to the energy that becomes available by nuclear fission (Szargut. 2005). Concerning radiative exergy, the ratio of the exergy flux to the total energy flux is 0.9336 at the Earth's surface (Szargut 2005).
Thermal exergy is represented by a carnot cycle to specify the obtainable work from a heat source (Szargut 2005) (Eq. 5):
2
1.3
Exergy of energy carriers
2.1
Exth = Qh * (Th – Tc) / Th
(5)
Exth : Thermal exergy (MJ) Qh : Energy from heat source (MJ) Th : Temperature of the heat source (K) Tc : Temperature of the environment (K) Extraction of thermal energy by air-water and brine-water heat pumps is considered in the corresponding datasets of ecoinvent data v1.2. In heat pumps, the temperature of the
184
Results Comparison of exergy factors to characterization factors from other methods
CExD is compared to CED (VDI 1997, Frischknecht et al. 2004), to the resource subcategories of EI'99 (Goedkoop et al.1999) and the CML'01 Method (Guinèe et al. 2001). The comparison is performed to identify differences in the relative weighting of resources. The number of assessed resources varies considerably between CExD, CED, EI'99 and CML'01. CExD provides exergy factors for 112 resources, while CED, EI'99 and CML'01 assess 12, 34 and 81 re-
Int J LCA 12 (3) 2007
LCA Methodology
Cumulative Exergy Demand
Table 2: Exergy of energy carriers in the ecoinvent database v1.2
rex-e (Ratio) b
Ex° (MJ/unit)
kg kg kg
Gross calorific value (MJ/unit) a 9.9 19.1 39.8
1.04 1.03 0.94
10.3 19.7 37.4
average fuel avg. of gasoline and lignite avg. of wood and lignite – – – – –
m3 kg kg kg m3 m3 m3 MJ
38.3 45.8 9.9 560,000 12,740 9,180 10,960 1
0.94 1.02 1.05 1.00 1.05 1.05 1.05 –
36.0 46.5 10.3 560,000 13,364 9,629 11,497 0
Energy, gross calorific value, wood average fuel in biomass Energy, kinetic, flow, in wind kinetic exergy velocity relative to surface Energy, potential, stock, potential exergy height relative to surface in barrage water Energy, solar radiative exergy – a Gross calorific values from Frischknecht et al. 2004; b Exergy to energy from Szargut (2005)
MJ
1
1.05
1.05
MJ MJ
1 1
1.00 1.00
1.00 1.00
MJ
1
0.93
0.93
Name ecoinvent
Coal, brown, in ground Coal, hard, unspecified, in ground Gas, mine, off-gas, process, coal mining Gas, natural, in ground Oil, crude, in ground Peat, in ground Uranium Wood, hard, standing Wood, soft, standing Wood, unspecified, standing Energy, geothermal
Name datasource exergy
Specification
Unit
lignite bitominous coal natural gas
average fuel average fuel average fuel
natural gas – – nuclear wood wood wood carnot cycle
sources, respectively. CED is an energy demand indicator and therefore considers only energetic resources. EI'99 and CML'01 exclude water consumption and all renewable resources, since they do not consider them to be exhaustible. The characterization factors of each resource in CExD, CED, EI'99 (H,A) and CML'01 are presented in Fig. 1. All factors were transformed into the unit crude oil-equivalents, to enable comparisons between the methods. The comparison between CExD and CED shows that the exergy and the energy content are very similar in energetic resources (R2>0.99 if only energetic resources are considered). This is due to the fact that most energetic resources in the ecoinvent database v1.2 feature a high quality and thus an exergy to energy ratio close to one. However, CED differs from CExD in that it does not consider non-energetic resources. The comparisons of CExD and EI'99 as well as between CExD and
CML'01 yield weak or no correlations (R2=0.2 and R2
